Auflistung nach:
Auflistung Band 45 - Heft 5 (Oktober 2022) nach Erscheinungsdatum
1 - 10 von 18
Treffer pro Seite
Sortieroptionen
- ZeitschriftenartikelCross domain fusion in power electronics dominated distribution grids(Informatik Spektrum: Vol. 45, No. 5, 2022) Sante, Pugliese; Landsiedel, Olaf; Kuprat, Johannes; Liserre, MarcoIn the near future, a drastic change in the structure of the electric grid is expected due to the increasing penetration of power electronics interfaced renewable energy sources (e.g. solar and wind), highly variable loads (e.g. electric vehicles and air conditioning) and unexpected energy demanding events (e.g. pandemics or natural disasters). Energy balancing management, voltage and frequency stability, reduced system inertia, grid resilience to fault conditions, and power quality of the supply are a few of the main challenges in the future power electronics dominated grids. Power electronics can solve these by integrating information and communication technology in new intelligent, highly reliable, and efficient devices like smart transformers. Smart transformers can increase the power flow flexibility by enabling the correct meshed-hybrid grid operations, as long as load mission and power generation profiles are known. Those profile are generally driven by heterogeneous, highly sparse and often incomplete data that belong to different domains. This article highlights the necessity of new approaches and models to identify patterns and events of interest that can serve as a common base. The resulting patterns can then be cross-fused in a common language and form the basis of further data analytics in future distribution grids.
- ZeitschriftenartikelAchtung.Datentrickserei(Informatik Spektrum: Vol. 45, No. 5, 2022) Lenz, Hans-J.
- ZeitschriftenartikelCross-domain fusion in smart seafloor sensor networks(Informatik Spektrum: Vol. 45, No. 5, 2022) Zainab, Tayyaba; Karstens, Jens; Landsiedel, OlafMany of the socio-economic and environmental challenges of the 21st century like the growing energy and food demand, rising sea levels and temperatures put stress on marine ecosystems and coastal populations. This requires a significant strengthening of our monitoring capacities for processes in the water column, at the seafloor and in the subsurface. However, present-day seafloor instruments and the required infrastructure to operate these are expensive and inaccessible. We envision a future Internet of Underwater Things, composed of small and cheap but intelligent underwater nodes. Each node will be equipped with sensing, communication, and computing capabilities. Building on distributed event detection and cross-domain data fusion, such an Internet of Underwater Things will enable new applications. In this paper, we argue that to make this vision a reality, we need new methodologies for resource-efficient and distributed cross-domain data fusion. Resource-efficient, distributed neural networks will serve as data-analytics pipelines to derive highly aggregated patterns of interest from raw data. These will serve as (1) a common base in time and space for fusion of heterogeneous data, and (2) be sufficiently small to be transmitted efficiently in resource-constrained settings.
- ZeitschriftenartikelCross Domain Fusion in der Archäologie – Interview mit Dr. Michael Kempf und Prof. Dr. Oliver Nakoinz(Informatik Spektrum: Vol. 45, No. 5, 2022) Renz, Matthias; Strohm, Steffen; Kempf, Michael; Nakoinz, Oliver
- ZeitschriftenartikelMitteilungen der GI im Informatik Spektrum 5/2022(Informatik Spektrum: Vol. 45, No. 5, 2022) null
- ZeitschriftenartikelMitteilungen der GI im Informatik Spektrum 4/2022(Informatik Spektrum: Vol. 45, No. 5, 2022) null
- ZeitschriftenartikelUm etliche Ecken ged8(Informatik Spektrum: Vol. 45, No. 5, 2022) Windenberg, Rolf
- ZeitschriftenartikelReinforcement learning as a basis for cross domain fusion of heterogeneous data(Informatik Spektrum: Vol. 45, No. 5, 2022) Christensen, Sören; Tomforde, SvenWe propose to establish a research direction based on Reinforcement Learning in the scope of Cross Domain Fusion. More precisely, we combine the algorithmic approach of evolutionary rule-based Reinforcement Learning with the efficiency and performance of Deep Reinforcement Learning, while simultaneously developing a sound mathematical foundation. A possible scenario is traffic control in urban regions.
- ZeitschriftenartikelGemini connector(Informatik Spektrum: Vol. 45, No. 5, 2022) Grossmann, Vasco; Nakath, David; Koch, Reinhard; Köser, KevinSpectacular advances have been made in the field of machine vision over the past decade. While this discipline is traditionally driven by geometric models, neural networks have proven to be superior in some applications and have significantly expanded the limits of what is possible. At the same time, conventional graphic models describe the relationship between images and the associated scene with textures and light in a physically realistic manner and are an important part of photogrammetry. Differential renderers combine these approaches by enabling gradient-based optimization in fixed structures of a graphics pipeline and thus adapt the learning process of neural networks. This fusion of formalized knowledge and machine learning motivates the idea of a modular differentiable renderer in which physical and statistical models can be recombined depending on the use case. We therefore present Gemini Connector: an initiative for the modular development and combination of differentiable physical models and neural networks. We examine opportunities and problems and motivate the idea with the extension of a differentiable rendering pipeline to include models of underwater optics for the analysis of deep sea images. Finally, we discuss use cases, especially within the Cross-Domain Fusion initiative.
- ZeitschriftenartikelCan neural networks predict steady annual cycles of marine ecosystems?(Informatik Spektrum: Vol. 45, No. 5, 2022) Slawig, Thomas; Pfeil, MarkusWe used artificial neural networks to replace the complete spin-up procedure that computes a steady annual cycle of a marine ecosystem driven by ocean transport. The networks took only the few biogeochemical model parameters and attempted to predict the spatially distributed concentrations of the ecosystem, in this case only nutrients, for one time point of the annual cycle. The ocean circulation was fixed for all parameters. Different network topologies, sparse networks, and hyperparameter optimization using a genetic algorithm were used. This showed that all studied networks can produce a distribution that is point-wise close to the original spin-up result. However, these predictions were far from being annually periodic, such that a subsequent spin-up was necessary. In this way, the overall runtime of the spin-up could be reduced by 13% on average. It is debatable whether this procedure is useful for the generation of initial values, or whether simpler methods can achieve faster convergence. Wir haben künstliche neuronale Netze verwendet, um den kompletten Spin-up zu ersetzen, mit dem ein stetiger Jahreszyklus eines marinen, durch den Ozeantransport angetriebenen Ökosystems berechnet wird. Die Netze nahmen nur die wenigen biogeochemischen Modellparameter und versuchten, die räumlich verteilten Konzentrationen des Ökosystems, hier nur Nährstoffe, für einen Zeitpunkt des Jahreszyklus vorherzusagen. Die Ozeanzirkulation war für alle Parameter fest. Es wurden verschiedene Netzwerktopologien, „sparse networks“ und ein Hyperparametertuning durch einen genetischen Algorithmus verwendet. Alle Netze konnten eine Verteilung erzeugen, die dem ursprünglichen Spin-up-Ergebnis punktweise ähnlich war. Allerdings waren die Vorhersagen weit davon entfernt, jahresperiodisch zu sein, weshalb ein nachträglicher Spin-up nötig war. So konnte die Gesamtlaufzeit des Spin-ups im Durchschnitt um 13 % reduziert werden. Es bleibt fraglich, ob dieses Verfahren sinnvoll ist, um Anfangswerte zu generieren, oder ob einfachere Methoden eine schnellere Konvergenz erreichen können.